Thrombin binding to fibrinogen at its substrate recognition sites in the central E domain leads to proteolytic conversion of fibrinogen to fibrin. Fibrin itself has an appreciable non-substrate thrombin binding potential termed 'anti-thrombin I', that is defined by two classes of thrombin-binding sites, one of low affinity in the E domain, and the other of high affinity in certain D domains. The low affinity sites represent a residual aspect of fibrinogen substrate binding, whereas the high affinity thrombin binding site in human fibrin(ogen) is situated exclusively in a gamma chain variant termed gamma', which contains a unique C-terminal sequence from residue 408 to 427 (y' l-427L) that also binds plasma factor XIII. The thrombin binding site on y' 1-427L encompasses residues 414 to 427, and Tyr sulfation at 418 and 422 as well as the ultimate tripeptide sequence (DDL), are important for thrombin binding. We believe that the most important physiological aspect of anti-thrombin I activity in blood is that non-substrate thrombin binding to fibrin sequesters thrombin within the forming clot, removing it from the thrombin-generating environment, thereby reducing thrombin feedback clotting activation as well as other direct thrombin effects. The goal of this proposal is to evaluate that hypothesis by investigating the role of non-substrate thrombin binding sites in murine fibrin that are representative of human anti-thrombin I activity. To accomplish this we will prepare genetically-altered mice in which (1) the non-thrombin-binding murine y' chain sequence has been replaced by the thrombin-binding human y' sequence (hu-y' fibrinogen), or (2) the low-affinity thrombin-binding site has been mutated at BB 68 to create fibrinogen with both defective thrombin substrate-recognition and low-affinity non-substrate thrombin-binding [Naples I (BB A68T)]. In Aim 1 we will characterize the biochemical properties of hu-y' fibrin(ogen), determine the in vivo consequences of introducing hu-y' on thrombosis, factor XIII-mediated crosslinking, and fibrinolysis, and carry out an interaction screen in thrombomodulin-deficient mice (TM pro, TM flox) to assess whether the hu-y' phenotype ameliorates or exacerbates the prethrombotic or flagrantly thrombotic state. We will also evaluate the content and in vivo metabolic conversion of intact y' 1-427L chains to non-thrombin-binding (des-EDDL) y' 1-423P chains, using ELISA methodology that has been developed for measuring human y' chains. In Aim 2 we will characterize the biochemical properties of murine Naples I fibrin(ogen), evaluate the in vivo consequences of introducing the Naples I mutation on thrombosis, and carry out an interaction screen in prethrombotic TM pro mice to determine whether the Naples I phenotype exacerbates the hypercoagulable state. We expect to resolve any controversy that now exists as to whether human y' chains have beneficial or deleterious in vivo effects on thrombosis or fibrinolysis. By comparing analyses on the hu-y' and Naples I phenotypes, we will be able to define the specific roles of the high- and low-affinity components of anti-thrombin I in the prevention and initiation of thrombosis.